Inspiring

Footage disappears when using VR Converter to convert 2D Source to Fisheye (Fulldome)

Forum|Forum|1 month ago
June 23, 2026
3 replies
77 views

Here are the steps to reproduce the problem:

Create a comp, 1024x1024
Create any 2D object, e. g. a text and center it in the comp
Add the VR Converter effect to the text layer
Set Input to 2D Source, set output to Fisheye (Fulldome)

Result: The text just disappears. Re-orienting the camera view does not do anything visible.

Expected result: The text is distorted for Fulldome and can be placed in the dome by re-orienting the camera view. This used to work in older versions of After Effects, as you can see for example in this Youtube tutorial at 1:35 :

My system: After Effects 26.3.0 (Build 87), WIndows 11, Nvidia RTX 5070 Ti, Driver 610.62

Henry_Collins

Participating Frequently

This sounds more like a regression in the newer AE builds than a project issue. I'd try the same setup in an older version to see if the behavior changes. If it's reproducible with those simple steps, it's probably worth filing a bug report with Adobe.

B

Bob9743Author

Inspiring

Thanks, that does work, even for higher resolutions. The center is of course the best starting point to then position the object on the dome - but of course 1 pixel does not make much of a difference especially with high resolutions, so I could at least go on using it. The behaviour is still odd though, especially when the center seems to be not a problem anymore once you changed the resolution.

Thanks,

Bob

Nishu Kushwaha

Community Manager

Glad to hear that helped. I’ll follow up with the team.

Thanks,

Nishu

Nishu Kushwaha

Community Manager

Hi Bob9743, I tested this on my end. I can confirm I am getting the same behavior. The interesting finding: it only reproduces when the composition is a perfect square (e.g., 1024x1024). If the comp dimensions are off by even a single pixel (e.g., 1024x1025), the VR Converter works as expected with the fisheye distortion. Before calling it a bug, I'd like to do some research on this internally.

As a quick workaround until this is fixed, you can change your comp to a non-square size (even something like 1024x1025), apply the VR Converter, and crop on export if you need a perfectly square final output.

I'll share as soon as I've an update.

Thanks,

Nishu

B

Bob9743Author

Inspiring

Hello NIshu,

thanks for looking at this. I can confirm, that this could be a workaround, but it is not a stable one: One pixel seems to work only for small resolutions. The usual output for Fulldome is 4096x4096. Changing this to 4097 does not work, 4100 seems to work again, maybe some other dimensions would work as well, I did not make a test run. What is also interesting to note: The VR Converter Effect changes the dimensions automatically back to a square when setting it to Fulldome, which basically makes sense, but makes the workaround even trickier to handle. It is even more confusing, when using several layers this way:

Start with 1024x1025
Create text, VR Converter, Fulldome → It stays visible, the comp has 1024x1024 now
Create another text, VR Converter, Fulldome → Text disappears, set comp to 1025 makes it appear again
Change the setting in one Converter effect output to equirectangular, then back to Fisheye → the comp has 1024x1024 again and the text stays visible

This has to be a bug, I can not think of a scenario, where this behaviour would be intentional.

Thanks,

Bob

Nishu Kushwaha

Community Manager

Hi Bob, thanks for the deeper testing.

During my own follow-up testing, I noticed something: the disappearance only happens when the text layer is perfectly centered in the comp. If you nudge the text even slightly off-center, it stays visible even in a perfectly square comp.

I'm still discussing this with engineering to confirm whether it's a known limitation or something they'd want to address.

In the meantime, the most reliable workaround is to offset your content by even a single pixel from the exact center.

Will follow up here as I have more from the team.

Thanks,

Nishu

Sign up

To post, reply, or follow discussions, please sign in with your Adobe ID.

Sign in to Adobe Community

To post, reply, or follow discussions, please sign in with your Adobe ID.